About sample rates during different stages of a song

Have an urge to learn, or a calling to teach? Want to share some useful Youtube videos? Do it here!
Post Reply
User avatar
sonicbyte
Posts: 347
Joined: 16 Jan 2015
Location: Argentina
Contact:

13 Jun 2017

Hi guys,

I've got a question, normally I compose my music on a 24/48khz session in Reason, basically all is generated inside Reason's devices, RE and now VST, I almost never record external audio (sometimes a couple of guitars, but is not the rule)... So I was wondering, once I've got all my arrangements done and I want to switch myself to "mixing mode", is ok if I switch my session to 24/96khz to get more headroom and then mastering at that quality? Or I must always start/end my session at the desired sample rate to get consistent results? My system isn't high-end at all so thats why I asking this. On the mixing stage I could export stems of all my instruments and get the most of my system in terms of DSP/RAM... so what do you think?

Btw, I don't want to open again the debate about if someone will hear any differences on 16/24 44.1/48/96k recordings and all that jazz... This is just to know your opinion about is more wise to start and end a song on the same sample rate or could get some better results if I'm able to switch session quality at different stages.

Thanks !

User avatar
QVprod
Moderator
Posts: 3488
Joined: 15 Jan 2015
Contact:

13 Jun 2017

Mixing at 96k will not give you more headroom nor make your mix better quality. Mixing at 24/48k is just fine. Even 44.1k is fine. The main difference is if you happen to have any synths that have aliasing issues at 48k, those aliasing issues would be eliminated. Aside from that, 96k will have no impact of the quality of the mix, but it will eat up dsp very quickly as you would be working the computer twice as hard.

User avatar
sonicbyte
Posts: 347
Joined: 16 Jan 2015
Location: Argentina
Contact:

13 Jun 2017

QVprod wrote:Mixing at 96k will not give you more headroom nor make your mix better quality. Mixing at 24/48k is just fine. Even 44.1k is fine. The main difference is if you happen to have any synths that have aliasing issues at 48k, those aliasing issues would be eliminated. Aside from that, 96k will have no impact of the quality of the mix, but it will eat up dsp very quickly as you would be working the computer twice as hard.
Thanks​ for the answer[emoji106]

Enviado desde mi XT1068 mediante Tapatalk

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

14 Jun 2017

QVprod wrote:Mixing at 96k will not give you more headroom nor make your mix better quality. Mixing at 24/48k is just fine. Even 44.1k is fine. The main difference is if you happen to have any synths that have aliasing issues at 48k, those aliasing issues would be eliminated. Aside from that, 96k will have no impact of the quality of the mix, but it will eat up dsp very quickly as you would be working the computer twice as hard.
There is one difference and has to do with the way compressors work - they tend to sound more like their analog counter-parts at higher sample rates according to some folks. Just read this from Bob Katz earlier today:

Bob Katz:
"Well, if I were working on a session that has new material to be integrated with samples from "ancient" samplers, I would upsample the samples to 3248 or 3296 (if 96k is your goal). YES, AGREED, this DOES NOT IMPROVE THE SOUND OF THE ORIGINAL SOURCE SAMPLES. BUT, IT does permit you to work the rest of the session at the higher sample rate, which audibly reduces the distortion of nonlinear calculations such as digital compression. Like chicken soup, working at the higher rate couldn't hurt, and it may help."

I guess the 'couldn't hurt' comment is based on having a kick ass computer where CPU limits are not a concern…. ;)


Sent from some crappy device using Tapatalk
Selig Audio, LLC

User avatar
QVprod
Moderator
Posts: 3488
Joined: 15 Jan 2015
Contact:

14 Jun 2017

selig wrote:
QVprod wrote:Mixing at 96k will not give you more headroom nor make your mix better quality. Mixing at 24/48k is just fine. Even 44.1k is fine. The main difference is if you happen to have any synths that have aliasing issues at 48k, those aliasing issues would be eliminated. Aside from that, 96k will have no impact of the quality of the mix, but it will eat up dsp very quickly as you would be working the computer twice as hard.
There is one difference and has to do with the way compressors work - they tend to sound more like their analog counter-parts at higher sample rates according to some folks. Just read this from Bob Katz earlier today:

Bob Katz:
"Well, if I were working on a session that has new material to be integrated with samples from "ancient" samplers, I would upsample the samples to 3248 or 3296 (if 96k is your goal). YES, AGREED, this DOES NOT IMPROVE THE SOUND OF THE ORIGINAL SOURCE SAMPLES. BUT, IT does permit you to work the rest of the session at the higher sample rate, which audibly reduces the distortion of nonlinear calculations such as digital compression. Like chicken soup, working at the higher rate couldn't hurt, and it may help."

I guess the 'couldn't hurt' comment is based on having a kick ass computer where CPU limits are not a concern…. ;)


Sent from some crappy device using Tapatalk
Interesting. I suppose how useful this is will depend on the the computer indeed. I'd be interested to know how noticable the difference between the two sample rates are.

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

14 Jun 2017

QVprod wrote:
selig wrote:
QVprod wrote:Mixing at 96k will not give you more headroom nor make your mix better quality. Mixing at 24/48k is just fine. Even 44.1k is fine. The main difference is if you happen to have any synths that have aliasing issues at 48k, those aliasing issues would be eliminated. Aside from that, 96k will have no impact of the quality of the mix, but it will eat up dsp very quickly as you would be working the computer twice as hard.
There is one difference and has to do with the way compressors work - they tend to sound more like their analog counter-parts at higher sample rates according to some folks. Just read this from Bob Katz earlier today:

Bob Katz:
"Well, if I were working on a session that has new material to be integrated with samples from "ancient" samplers, I would upsample the samples to 3248 or 3296 (if 96k is your goal). YES, AGREED, this DOES NOT IMPROVE THE SOUND OF THE ORIGINAL SOURCE SAMPLES. BUT, IT does permit you to work the rest of the session at the higher sample rate, which audibly reduces the distortion of nonlinear calculations such as digital compression. Like chicken soup, working at the higher rate couldn't hurt, and it may help."

I guess the 'couldn't hurt' comment is based on having a kick ass computer where CPU limits are not a concern…. ;)


Sent from some crappy device using Tapatalk
Interesting. I suppose how useful this is will depend on the the computer indeed. I'd be interested to know how noticable the difference between the two sample rates are.
Indeed - It's been a while since I did any comparative listening, but I struggled to hear any discernible distortion with compression @ 44.1 vs 96 kHz sample rates. Could be different with different compressors, or something only the golden ears mastering engineers can hear! Also, not all distortion is bad - maybe I just got used to the distortion at 44.1 kHz, and that made it difficult for me to hear it as "distortion"?


Sent from some crappy device using Tapatalk
Selig Audio, LLC

User avatar
QVprod
Moderator
Posts: 3488
Joined: 15 Jan 2015
Contact:

14 Jun 2017

selig wrote:
QVprod wrote:
selig wrote:
QVprod wrote:Mixing at 96k will not give you more headroom nor make your mix better quality. Mixing at 24/48k is just fine. Even 44.1k is fine. The main difference is if you happen to have any synths that have aliasing issues at 48k, those aliasing issues would be eliminated. Aside from that, 96k will have no impact of the quality of the mix, but it will eat up dsp very quickly as you would be working the computer twice as hard.
There is one difference and has to do with the way compressors work - they tend to sound more like their analog counter-parts at higher sample rates according to some folks. Just read this from Bob Katz earlier today:

Bob Katz:
"Well, if I were working on a session that has new material to be integrated with samples from "ancient" samplers, I would upsample the samples to 3248 or 3296 (if 96k is your goal). YES, AGREED, this DOES NOT IMPROVE THE SOUND OF THE ORIGINAL SOURCE SAMPLES. BUT, IT does permit you to work the rest of the session at the higher sample rate, which audibly reduces the distortion of nonlinear calculations such as digital compression. Like chicken soup, working at the higher rate couldn't hurt, and it may help."

I guess the 'couldn't hurt' comment is based on having a kick ass computer where CPU limits are not a concern…. ;)


Sent from some crappy device using Tapatalk
Interesting. I suppose how useful this is will depend on the the computer indeed. I'd be interested to know how noticable the difference between the two sample rates are.
Indeed - It's been a while since I did any comparative listening, but I struggled to hear any discernible distortion with compression @ 44.1 vs 96 kHz sample rates. Could be different with different compressors, or something only the golden ears mastering engineers can hear! Also, not all distortion is bad - maybe I just got used to the distortion at 44.1 kHz, and that made it difficult for me to hear it as "distortion"?


Sent from some crappy device using Tapatalk
Perhaps, back when I ran a studio (I sound old, it was only 3 years ago haha), had a coworker who ran everything at 96k while I stuck to 44.1 and 48k. Two of the songs I mixed for a project were recorded by him at 96k and I definitely never noticed a difference. Could be my ears though.

Post Reply
  • Information
  • Who is online

    Users browsing this forum: No registered users and 11 guests